5 papers
Weak-to-Strong Generalization via Direct On-Policy Distillation
Shiyuan Feng, Huan-ang Gao, Haohan Chi +7
Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because t…
Spectral Rewiring for Exploration, Purification, and Model Merging
Zhilong Zhang, Hongli Yu, Huan-ang Gao +5
Reinforcement learning has become a standard post-training recipe for large language models, but dense full-parameter updates create two deployment-relevant bottlenecks: suppressed…
DSPO: Stable and Efficient Policy Optimization for Agentic Search and Reasoning
Chenyang Gu, Yewen Pu, Bruce Yang +2
Enhancing LLMs with the ability to actively search external knowledge is crucial for complex and real-world tasks. Current approaches either rely on prompting to elicit the model's…
CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
Weinan Dai, Hanlin Wu, Qiying Yu +13
GPU kernel optimization is fundamental to modern deep learning but remains a highly specialized task requiring deep hardware expertise. Despite strong performance in general progra…
CodeAgents: A Token-Efficient Framework for Codified Multi-Agent Reasoning in LLMs
Bruce Yang, Xinfeng He, Huan Gao +3
Effective prompt design is essential for improving the planning capabilities of large language model (LLM)-driven agents. However, existing structured prompting strategies are typi…